Presentation on theme: "1 RLINE: A Line Source Dispersion Model for Near-Surface Releases Presented at the 12 th Annual CMAS Conference, Chapel Hill, NC October 28 – 30, 2013."— Presentation transcript:
1 RLINE: A Line Source Dispersion Model for Near-Surface Releases Presented at the 12 th Annual CMAS Conference, Chapel Hill, NC October 28 – 30, 2013 Saravanan Arunachalam 1, Michelle G. Snyder 2, Akula Venkatram 3, David K. Heist 2, Steven G. Perry 2, Vlad Isakov 2 1 UNC/IE, 2 U.S.EPA/NERL/AMAD, 3 UCR
2 Source: Rowangould, Transport Research Part D, 2013 Background 59.5 million people live within 500m of roads with > 25,000 Annual Average Daily Traffic (AADT)
3 Illustrative AADT in Urban Area Source: Michigan DOT
4 Why was RLINE Developed? Based on Science Advisory Board and the National Research Council recommendations, EPA-ORD initiated research on near-road air quality and health effects Field measurements indicated that exposures to traffic- emitted air pollutants near roads can be influenced by complexities of roadway configurations (noise walls, depressed sections, etc.) To improve exposure metrics for future health studies, ORD initiated model development project to account for these important parameters
5 What is RLINE? RLINE is EPA -ORD's research dispersion modeling tool for near roadway assessments based on a steady-state Gaussian formulation currently formulated for near-surface releases contains new formulations for the vertical and lateral dispersion rates accounts for low wind meander includes M-O similarity profiling of winds near the surface uses the surface meteorology provided by AERMET includes user-friendly input requirements for road network
6 EPA-ORD: Model Development Focus Account for near-road complexities and very near road concentrations (i.e., within a few meters of the road) Design and conduct wind tunnel and field studies for development and evaluation of improved line source algorithms RLINE is initial modeling product of this development program research tool designed primarily to support risk assessments and health studies related to near-road pollutants
7 What RLINE is NOT? RLINE is not designed for regulatory applications (e.g., NAAQS enforcement, New Source Review, PM Hot Spot Conformity, SIP analyses, etc). It is contained in a research platform and has not gone through the rigorous review and public comment required for inclusion in the list of recommended regulatory models
8 Model Algorithm Steady-state, Gaussian based plume model Concentration at a receptor is the integrated contributions from points along the line using a Romberg Iteration Scheme. where and For each point source within the integration, a wind-direction following plume is simulated. Plume For each point source within the integration, meander is handled in the same way as AERMOD point source meander. Meander Snyder et al, Atmos. Environ, 2013
9 Model Algorithm RLINE uses the meteorological inputs from the AERMOD met. preprocessor (u *, w *, mixing height, L mo, z 0, Wspd, Wdir, heat flux, T) then corrects u * for light wind conditions. Other surface parameters are then recalculated to be internally consistent with M-O theory. This method was developed and evaluated by Qian and Venkatram (BLM 2011). Meteorological Inputs Concentration is a function of mean plume height (similar in nature to the effective parameter concept used in AERMOD). Mean Plume Height Snyder et al, Atmos. Environ, 2013
10 Model Development Databases Tracer Studies o Idaho Falls 2008 o Prairie Grass Wind Tunnel Studies Field Studies o Raleigh, Detroit, Phoenix
11 How is RLINE being used? The model can support health and risk assessments, epidemiology studies, and community based tools. RLINE is designed to handle urban situations with thousands of line sources and receptors. RLINE is used in a) “Near-Road EXposures to Urban air pollutants Study” (NEXUS) in Detroit examining the role of near-road exposures on asthmatic children who live near major roadways; b) Portland, ME and c) North Carolina Piedmont RLINE can be used to estimate spatial variability in urban areas with many roadways. See Talks by Snyder et al, and Chang et al, and Poster by Isakov et al.
12 How does RLINE support community tools? R-LINE algorithm is used in C-LINE, a decision support tool for evaluating effects of alternate transportation options on community health EPA’s C-LINE Tool See Poster by Barzyk et al
13 Ongoing Development Work The model framework is designed to accommodate algorithms for simulating near-source effects of complex roadway configurations (noise barriers, depressed roadways, roadway vegetation, etc). These configurations could be used in mitigation, to reduce exposure near high emission roadways. Finn et al., Atmos. Environ., 2010 _____ no barrier barrier Idaho Falls 2008 tracer study
14 Field and wind tunnel studies show noise barriers and depressions reduce downwind concentrations under multiple stability conditions. These features are being added to the flat terrain RLINE model. Field studies Ongoing Development Work
15 4 models applied for 2 Tracer Studies Idaho Falls 2008 CalTrans Highway 99 All models showed ability to estimate majority of observations within a factor of 2 Models performed best for near-neutral conditions in both tracer studies Mixed results in convective and stable conditions Inter-comparison against other models Heist et al, Trans. Res. D, 2013 AERMOD-A AERMOD-V
16 Uses hourly meteorological data from AERMET v12345 Similar to AERMOD applications Contains updated σ y and σ z formulations See Venkatram et al, 2013 Can handle inputs in the form of activity (in AADT) or emissions (in g/m/s) AADT outputs can be post-processed with TAFs and species- specific emissions to obtain variable emissions Output in μg/m 3 Produces grouped source outputs Hourly (for entire simulation period or monthly) Daily average Features in Release Version of RLINE
17 Analytical solution for near-source receptors Speeds up computation Input # lanes per roadway Spreads emissions over the lanes of traffic Solid/noise barrier algorithm Depressed roadway algorithm Features Available as Beta Options
18 Summer 2013: Availability of pre-release version announced via Community Modeling & Analysis System (CMAS) website November 2013: Full public release expected via CMAS to include: Explanation of model code, use, and inputs required Documented evaluation and evaluation data Qualifications on appropriate use Spring/Summer 2014: Planned repository of tools and utilities for pre- and post-processing to support user community Where to get RLINE from?
19 Disclaimer: Although this work was reviewed by EPA and approved for publication, it may not necessarily reflect official Agency policy.
20 Backup Slides
21 Model Algorithm New Dispersion Formulations Vertical Dispersion Venkatram et al, Atmos. Environ, 2013
22 Lateral Dispersion Model Algorithm New Dispersion Formulations Prairie Grass Venkatram et al, Atmos. Environ, 2013
23 Wind tunnel studies Baldauf et al, 2009 Flat terrain Downwind barrier Depressed Roadway Heist et al, Atmos. Environ, 2009 Ongoing Development Work H = 6 m